Amazon Redshift now introduces Late Materialization to improve performance for queries

Amazon Redshift now utilizes the late materialization to lessen the quantity of data scanned and enhanced the performance of queries with the implied filters. Late materialization row-level filtering decreases I/O for queries with filters by factoring and batching in the filtering of predicate before searching data blocks in the next column. Amazon Redshift with late materialization fetches a group of data CUSTOMER_STATUS_LEVEL and CUSTOMER_SINCE_DATE then implement the particular predicates. If the 10 percent of the CUSTOMER_DETAIL table rows content the predicate filters so as a result, Amazon Redshift can possibly save 90 percent of the I/O for the remaining columns that ultimately improves the query performance.